is AV= fair in matchpoints?

[Mirjam_3]

New as td i started to use A== if time is up as i see most other td's do (though i saw one giving A+)... the more i am on this site, the more i wonder whether this is a correct adjust.

 

In imps it is ok, since 0 imps is nothing... but in matchpoints i start to feel this is injust if it is not the players fault they are late... If the player ends up with a 60% score the 50% will put him/her down...

 

Could u tell me yr opinion plse?

[inquiry]

It is a matter of overall fairness.

 

Let's start with the issue of time. Bridge is a timed event. To award average plus to a table/partnership not finishing on time is not exactly right. The duty of the players involved is to try to finish on time. Of course, someone dragging geet to get average minus or average need to be dealt with.

 

Second issue, if you award an average plus to a pair, you "hurt" all the other pairs in the field. If you award an average, you hurt only one pair. That is a philosophical hurt of course.

 

ben

[Mirjam_3]

I see, Ben. And what if it is the fault of the td?

 

Today i subbed in a tournament and the situation was i got in the place of a player who was there. My partner counted time untill it was gone... When he started there were 4 minutes to go... the td didnot manage (?) to get the real disconnected player subbed so play was not possible. We got av=.

Maybe this is a bad example, but it happened...

 

I also see as td that players really bag me to come and watch the table when time is running out since most play faster if i am there.. and they want to avoid the av=.

 

I agree about yr second point though. To give av+ wd hurt the scores...

 

Isnot there any other solution possbile? Like let the board not count? Or do u think that wd not be ok either?

[Cascade]

This is a difficult area for online bridge.

 

What I like to do is award an adjusted score equal to the likely table result.

 

What I would like to do is assess a procedural penalty on top of that when I judge that a particular pair has unnecessarily caused the slowness.

 

Av+ Av- and Av= almost always distort the other scores.

 

The model of face to face bridge is that pairs that finish late complete their board and are expected to catch up on subsequent rounds. If they do not then it is possible for the director to warn and assess penalties for slow play. In my experience as a face to face director I have issued many warnings and very few penalties. People cooperate if they know you are going to harm them.

 

In theory at least it would be easy to get good information about who was using all of the time in online bridge and therefore to make good judgements in assessing the blame for the slow play.

 

At times I remove a board from a table when they are so far behind that they are very unlikely to finish it during the round. This is a bit like Uday's no new boards when there is less than 4 minutes on the clock (or whatever the number is). In this case I just give both pairs no score on this board - that is it does not count towards their score. If I thought a pair was deliberately doing this to protect their good score or for some other reason then I would issue them a penalty. I would not be lenient with this sort of penalty.

 

If there is a normal (non-Swiss) movement then flexibility in the finishing time for a round could be possible. So the round could end with a few tables still playing. These tables would be expected to finish quickly and then to catch up on the next round. They would also be liable for slow play penalties. It could be possible for these penalties to be assessed automatically. Warning for 1st offence less than x seconds overtime, penalty for second offence, bigger penalty for 3rd offence etc.

 

In a Swiss event we need all tables finished before assigning tables for the next round. One solution that is used in face to face bridge is that assignments for round n are based on the results after round n-2. If this was done then the same procedures outlined above could be used. Alternatively heavy penalties for slow play preferrably based on accurate information combined with the removal of boards. In extreme cases noncompliance could result in removal from the tournament - with substitutes it would be possible to do this without affecting any other pair in the event.

 

Another alternative for Swiss would be to make the assignments based on incomplete results. With slow players moving when they are finished and being expected to catch-up under the threat of penalties.

 

In my experience being expected to catch up under the threat of penalties works well for face-to-face bridge.

 

To make the tournament director's job easier I think we need the flexibility to assess procedural and disciplinary penalties.

 

I also think that some flexibility in the way that the transition is made between rounds would be helpful.

 

I imagine penalties and more flexible artificial assigned scores would be relatively easy to program. I am not sure how easy it would be to program more flexible transition from one round to the next.

[Gerardo]

BBO average calculations at matchpoints is somewhat flawed.

 

Ave+ is the better of 60% and your average.

Ave- is the worse of 40% and your average.

Ave is not 50%, it is what it says, your average. Your global average is not affected by the inability to play the hand.

 

It needs to distinguish hands actually played from hands which are averaged. There are a third group, hands adjusted to something not average. Not sure in which of the other groups these belong.

 

I think these belong to 1st group, but if not, things could get complicated. For example, director adjusting wrong hand (been there, done that). In this case, I suggest to keep the table score, allow it to be overruled by the adjusted score, and make a "table" result to blank the adjusted score, making the table score current.

 

If not, a simpler "adjusted?" flag should be enough(?)

[hrothgar]

Please note that this recommendation is NOT grounded in the Laws of bridge. The primary intent of the Laws is to restore equity. I prefer to implement a "solution" design to incentivise players to solve the problem.

 

Assume for the moment, that players run out of time in the middle of a board.

Furthermore, a late play is not feasible.

 

The best possible solution is for the director to study the board and and to assign the score that the players would have recieved had the board been played to completion. Unfortunately, this is not always practical.

 

I suggest the following:

 

In theory, the BBO server could track how much time each player has used during the course of a round. [i'm not sure whether this feature has been implemented or not]. Calculate the total amount of time that each side used during the bidding/play for the round in question. Note that the round incorporates multiple boards. Use this ratio to allocate the matchpoint allocations for the fouled board.

 

The are clear problems with this approach. Most notably, this has the potential to hand a pair an undeserved A+ or some such which has the potential to skew the results of the tournament. However, this has the following advantages:

 

1. Players are left with a clear incentive to make efficient use of time. (this assumes that players understand the hand adjustment algorithm)

2. The algorithm is simple to understand and objective.

2. Directors do not have to make coulda / woulda / shoulda decisions to manually adjust hands.

 

I think that the benefits outweigh the costs.

[Gerardo]

I reviewed relevant Laws:

 

Law 12 Director's discretionary powers

C. Awarding an Adjusted Score

1. Artificial Score

When, owing to an irregularity, no result can be obtained, the Director awards an artificial adjusted score according to responsibility for the irregularity: average minus (at most 40% of the available matchpoints in pairs) to a contestant directly at fault; average (50% in pairs) to a contestant only partially at fault; average plus (at least 60% in pairs) to a contestant in no way at fault (see Law 86 for team play or Law 88 for pairs play). The scores awarded to the two sides need not balance.

 

LAW 88 - AWARD OF INDEMNITY POINTS

In a pair or individual event, when a non-offending contestant is required to take an artificial adjusted score through no fault or choice of his own, such contestant shall be awarded a minimum of 60% of the matchpoints available to him on that board, or the percentage of matchpoints he earned on boards actually played during the session if that percentage was greater than 60%.

 

So, I was wrong, Ave is 50% at pairs

and BBO assumes both are partially at fault when time is up and round is not over. But both parties, particularly the non-offending one, should call director ASAP when time issues show up.

 

Richard, your scenario need client reporting actual time taken on a bid/play, to avoid charging lag.

Would be very, very useful.